Comparison of dextran and albumin on blood coagulation in patients undergoing major gynaecological surgery

Johann Sigurjonsson, David Hedman, Peter Bansch, Ulf Schött

Hydroxyethyl starches have been withdrawn from the European market. In Sweden, dextran was the main colloid until 2000, when starches overtook the market. After the recent 6S-trial, it was suggested that dextran could be reinstituted, but concerns for greater coagulopathy, bleeding and anaphylaxis still remain. An experimental study from our department indicated that isovolemic substitution of dextran-70 did not derange the von Willebrand function more than albumin 5%, considering the fact that dextran is hyperoncotic in comparison to albumin 5% and, therefore, induces a greater plasma volume expansion and thereby a greater dilutional coagulopathy. Eighteen patients undergoing major gynaecological surgery were assigned to receive either 5% albumin or 6% dextran-70 with 9 patients in each group. Standard coagulation tests, including prothrombin time (PT), activated partial thromboplastin time (aPTT), fibrinogen and platelet count, viscoelastic coagulation test thromboelastometry (ROTEM) and the Multiplate platelet aggregation test were used to test for coagulation defects at different time points perioperatively. Blood loss, blood loss replacement data and haemodynamic parameters were retrieved from anaesthetic and postoperative charts. A local departmental fluid and transfusion/infusion protocol assured haemoglobin > 90 g/l and mean arterial pressure > 65 mmHg with Ringer's acetate in addition to the colloid use. There were no differences in demographic data between the groups. The tissue factor-activated (EXTEM) clot-structure parameter ROTEM A10 was decreased significantly in the dextran group as compared to the albumin group after the infusion of 500 ml of either colloid solution. The PT and aPTT were significantly prolonged, and the platelet count decreased postoperatively in the dextran group, whereas albumin only deranged fibrinogen levels as compared to preoperative levels. There were no differences in Multiplate platelet aggregometry, amount of haemorrhage or transfusion of blood components between the groups. Standard plasma-based coagulation tests, platelet count and whole blood viscoelastic clot structure are affected by 6% dextran-70 to a greater extent than by 5% albumin, but platelet aggregation is not. Future studies should use more advanced haemodynamic monitoring to assess isovolemic plasma volume expansion with dextran and whether this affects haemostasis to a lesser degree.